电网调度防误人工智能模型的构建与操作分类方法
DOI:
CSTR:
作者:
作者单位:

贵州电网有限责任公司电力调度控制中心

作者简介:

通讯作者:

中图分类号:

基金项目:

基于深度学习的电网调度全业务智能防误技术研究及应用(GZKJXM20220105)


Construction of Artificial Intelligence Model for Misoperation Prevention ofPower Grid Dispatching and Operation Classification
Author:
Affiliation:

Fund Project:

  • 摘要
  • |
  • 图/表
  • |
  • 访问统计
  • |
  • 参考文献
  • |
  • 相似文献
  • |
  • 引证文献
  • |
  • 资源附件
  • |
  • 文章评论
    摘要:

    随着电网规模的逐年增长与复杂程度的提高,电网调度操作越来越频繁,任何操作失误都可能导致严重后果。针对上述挑战,本文提出一种电网调度防误人工智能模型的构建与操作分类方法。首先,结合卷积神经网络(Convolutional Neural Network, CNN)和长短期记忆网络(Long Short-Term Memory, LSTM)的优点,构建基于CNN-LSTM的电网调度防误模型,以深入挖掘电网调度操作的深层特征和时间序列特性,提升误操作识别的精准度,增强电网调度的防误效率。其次,将决策树算法与布尔公式、贝叶斯公式结合,提出基于C4.5决策树的误操作分类方法,实现精细化的误操作分类。仿真结果表明,相比于对比算法,所提算法将防误成功率分别提高了2.50%和3.29%。

    Abstract:

    With the increasing scale and complexity of the power grid year by year, power grid dispatching operations are becoming more frequent, and any misoperations can lead to serious consequences. In response to the above challenges, this article proposes a construction of artificial intelligence model for misoperation prevention of power grid dispatching and operation classification. Firstly, by combining the advantages of convolutional neural network (CNN) and long short-term memory (LSTM), a CNN-LSTM based misoperation prevention model for power grid dispatching is constructed to explore the deep and temporal characteristics of power grid dispatching operations, achieve more accurate identification of misoperations, and improve the efficiency of power grid dispatching misoperation prevention. Secondly, combining the decision tree algorithm with Boolean and Bayesian formulas, a C4.5 decision tree based misoperation classification method is proposed to achieve refined misoperation classification. The simulation results show that compared to the comparative algorithms, the proposed algorithm increases the success rate of misoperation prevention by 2.50% and 3.29% separately.

    参考文献
    相似文献
    引证文献
引用本文
分享
相关视频

文章指标
  • 点击次数:
  • 下载次数:
  • HTML阅读次数:
  • 引用次数:
历史
  • 收稿日期:2024-11-08
  • 最后修改日期:2024-11-08
  • 录用日期:2026-07-08
  • 在线发布日期:
  • 出版日期:
文章二维码